Orientation in Conspiration

The proposed project Orientation in Conspiration (ORION) aims to better understand the different factors that lead to the popularity of conspiracy theories (CTs) in online environments. CTs have been regarded as harmless curiosity for the better part of the last century. Nevertheless, the rapid spread of anti-vaccine misinformation in connection to the COVID-19 pandemic that was carried to a significant amount by online conspiracy groups has exemplified the concrete harms that belief in CTs can cause. The existence of CTs is in itself not purely bad: Conspiracy believers can fulfil a social role as watchdog, forcing social elites to justify their actions. Nevertheless, the intersection of CTs with the rapid spreading potential in online environments and coordinated efforts by outside actors to increase their popularity has tipped the balance towards causing more harm than good for society. Even though most policy makers agree that the spread of CTs in online environments should be curbed, intervening in such environments without massive censorship or the stigmatisation of individual conspiracy believers is an unsolved challenge. In ORION, we conceptualise the spread of CTs as a phenomenon that is driven by various individual, social and technological aspects of online environments. We aim to understand the different contributions of these aspects to the spread of CTs with the aim of informing interventions that target platform design and algorithms, and activities that are coordinated by outside actors rather than stigmatising and censoring individual conspiracy believers. It is high time to provide empirical evidence that can support such interventions, since the alternatives would mean massive infringements of user privacy and free speech, endangering the internet as we know it. Regarding CTs, the German language sphere is particular and different from other cultural contexts. This is on the one side driven by the historical context and high prevalence of right-wing nationalism, and on the other hand by widespread belief in esotericism and alternative medicine. Since we are particularly interested in contributing insights to policy making efforts in Austria, we focus on CTs that spread in German-language Telegram channels. We make use of a publicly available dump of these channels (the Schwurbelarchiv) that was made available via the Internet Archive last July. This data set is exceedingly large (24TB) and hard to work with, as the data is structured as HTML files and a large fraction of it is multimedia content. A significant contribution of ORION will therefore be to clean and structure the data set and transcribe significant amounts of the multimedia content to enable the use of advanced natural language processing approaches. We will make this clean and enriched version of the data set including a detailed data set descriptor available in a FAIR manner, with the explicit intent of enabling other research groups to build on it.